Genome-wide association studies (GWAS) have identified hundreds of risk variants associated to common diseases and other traits, yet in most cases have explained only a small fraction of genetic heritability. The source of this missing heritability is a critical question in human genetics. Efforts to address this question have largely focused on the search for specific disease risk variants, leaving the bulk of heritability still unexplained. We will systematically deconstruct genetic heritability by investigating how genetic content shared either via segments inherited identical-by-descent (IBD) or genotyped alleles identical-by-state (IBS), either genome-wide or partitioned across the genome, corresponds to phenotypic similarity across a broad range of human traits. We will use IBD to characterize the set of all risk variants, and IBS to characterize the subset of risk variants genotyped or tagged by GWAS chips. We will explore the idea that genetic similarity between unrelated individuals can be as useful, or more useful, than similarity between close relatives in quantifying and understanding genetic heritability. Partitioning heritability by genomic location will enable us to draw conclusions about the genetic architecture of various human traits, considering both the set of all risk variants and the subset of risk variants captured by GWAS chips. We will develop these ideas via simulation and apply them to real GWAS data comprising 70,000 samples. We will continue to release practical, publicly available software packages implementing the methods that we develop.